[U-Boot] [PATCH] MX5: Reuse the gd->tbl value for timestamp and add gd->lastinc for lastinc bss

The usage of bss values in drivers before initialisation of bss is forbidden. In that special case some data in .rel.dyn gets corrupted.
This patch is the same as recently applied for arm926js architecture.
Signed-off-by: Stefano Babic sbabic@denx.de CC: Heiko Schocher hs@denx.de --- arch/arm/cpu/armv7/mx5/timer.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/arm/cpu/armv7/mx5/timer.c b/arch/arm/cpu/armv7/mx5/timer.c index 3044fcf..1972f64 100644 --- a/arch/arm/cpu/armv7/mx5/timer.c +++ b/arch/arm/cpu/armv7/mx5/timer.c @@ -44,8 +44,10 @@ static struct mxc_gpt *cur_gpt = (struct mxc_gpt *)GPT1_BASE_ADDR; #define GPTCR_CLKSOURCE_32 (4<<6) /* Clock source */ #define GPTCR_TEN (1) /* Timer enable */
-static ulong timestamp; -static ulong lastinc; +DECLARE_GLOBAL_DATA_PTR; + +#define timestamp (gd->tbl) +#define lastinc (gd->lastinc)
int timer_init(void) {

Dear Stefano Babic,
In message 1295798910-20382-1-git-send-email-sbabic@denx.de you wrote:
The usage of bss values in drivers before initialisation of bss is forbidden. In that special case some data in .rel.dyn gets corrupted.
This patch is the same as recently applied for arm926js architecture.
Signed-off-by: Stefano Babic sbabic@denx.de CC: Heiko Schocher hs@denx.de
arch/arm/cpu/armv7/mx5/timer.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/arm/cpu/armv7/mx5/timer.c b/arch/arm/cpu/armv7/mx5/timer.c index 3044fcf..1972f64 100644 --- a/arch/arm/cpu/armv7/mx5/timer.c +++ b/arch/arm/cpu/armv7/mx5/timer.c @@ -44,8 +44,10 @@ static struct mxc_gpt *cur_gpt = (struct mxc_gpt *)GPT1_BASE_ADDR; #define GPTCR_CLKSOURCE_32 (4<<6) /* Clock source */ #define GPTCR_TEN (1) /* Timer enable */
-static ulong timestamp; -static ulong lastinc; +DECLARE_GLOBAL_DATA_PTR;
+#define timestamp (gd->tbl) +#define lastinc (gd->lastinc)
Why do we need these parens?
Best regards,
Wolfgang Denk

On 01/23/2011 05:28 PM, Wolfgang Denk wrote:
Why do we need these parens?
Checkpatch raises warnings without them.
Best regards, Stefano Babic
participants (2)
-
Stefano Babic
-
Wolfgang Denk